Journal of Economics and Sustainable Development www.iiste.

org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
25
Foreign Aid, Domestic Revenue and Economic Growth in Ghana

Joseph Kwadwo Tuffour (PhD)
School of Research and Graduate Studies; University of Professional Studies, Accra
jktuffour@yahoo.co.uk

Abstract
Various funding sources are available to any government in a developing country. In spite of their availability,
each source has its unique way of contributing to economic growth. The objective of this study is to determine
the relationship between foreign aid and domestic revenue on one hand and their effect on economic growth in
Ghana. Using macroeconomic time series data (for 1970 2011) and error correction methodology together with
Granger causality tests, the results show that, domestic revenue and foreign aid complement each other for
development financing. Moreover, foreign aid cannot be a substitute for domestic revenue generation as
domestic revenue is the most important between the two factors. Domestic revenue is the most effective
irrespective of the existence of other forms of capital. In addition, the threshold effect of aid becomes
non-existent when there is no capital accumulation. There is a positive causal link from both domestic revenue
and foreign aid to economic growth. It is recommended that, more attention should be given to the generation of
domestic revenue.
Key Words: Foreign Aid, Domestic Revenue, Economic Growth, Error Correction Model, Ghana

1.0 Introduction
Over the years, the quest to improve economic growth and reduce poverty has taken centre stage in most
developing countries. This requires massive expenditure, with resources obtained from external loans, grants and
domestic resource mobilisation. Available data shows that, the flow of capital in Ghana has continually increased.
The expectation is that, the increasing financial resources would translate into enhanced economic growth and
poverty reduction. However, this expectation is not met and Ghana has not experience accelerated economic
growth. This raises the question of how has external aid, Foreign Direct Investment (FDI) and domestic
resources contributed to enhanced economic growth and poverty reduction in Ghana? It is against this
background that this study investigates the external aid financing of economic growth in the Ghanaian context.
Despite the improvements in economic performance, Ghana continuous to be confronted with a number of
constraints, including levels of savings and investment that are low to fuel the growth needed to raise living
standards, reduce poverty and generate sufficient productive employment, in other words, for self-sustained
growth. Among the sources of funding available are foreign aid and domestic revenue. An understanding of aid
and domestic revenue patterns, and how they are related to economic growth and poverty reduction is thus, one
of the key elements of policy that fosters development of low income countries such as Ghana. The broad
objective of this study is to empirically estimate the impact of foreign aid and domestic revenue on economic
growth in Ghana. Specifically, this is to: identify whether Ghanas receipt of aid has significant effect on
economic growth; determine the role of domestic revenue in output enhancement; and determine whether or not
the flow of aid influences domestic revenue generation. The rest of the paper is organised as follows. The
background of foreign aid, domestic revenue and economic growth are in section two. Section three contains the
methodology and model specification. The data analysis and results are in section four while the conclusion is in
section five.

2.0 Background Issues
After many years of development assistance to developing countries, there is evidence that millions live on less
than US$1 per day as extreme poverty and others on US$2 per day, which is estimated to be about 40% of the
worlds population (UNDP, 2007). In addition to the inability to reduce poverty, economic growth in
Sub-Saharan Africa for instance has not seen much acceleration. The earlier view that low savings or low capital
formation is the cause of the start of vicious cycle of poverty and low growth is thus questionable. Given these
conditions, how has aid impacted on economic growth in Ghana is vital for domestic revenue mobilisation and
growth. The economic growth rate of Ghana since independence has been fluctuating, from 1970 to 1983 (see
Figure 1). Basically, this period witnessed negative growth rates over the history of Ghana. Ironically, the
negative growth rates coincided with military interventions in governance in the country. From 1979, the growth
rate in the country started to decline until the intervention of the SAP. Since the establishment of the SAP, Ghana
has recorded positive growth rates till date. However, the positive trend has been relatively stable but not
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
26
accelerating, to lead to massive poverty reduction (see Figure 1 below). It is therefore necessary for Ghana to
intensify effort to improve upon the fairly stable growth rate.


In a similar way, domestic revenue has not seen massive growth, although there is an upward trend. It is
observed that, there were large swings in the growth of domestic revenue generation from 1970 to 1980 (see
Figure 2). The low revenue resources during the early periods of the 1980s are an indication of the general
decline in the Ghanaian economy. At the same period, there were low aid flows to Ghana. With the advent of the
SAP, the growth of domestic revenue picked up and the trend has been positive since then. But this needs to
accelerate in order to enhance public expenditure. Also, after the SAP, aid has improved in magnitude, but it is
below that of domestic revenue. This is expected as aid is supposed to complement domestic resources but not
become the main source of a countrys financing. Comparatively, the uncertainty about aid flows is clearly seen
in Ghana as there are fluctuations even after the SAP. Thus, it is expected that, the larger the gap between
domestic revenue and aid, the more the effort required to generate more domestic revenue.


Historically, Ghana has extensively depended on both domestic and foreign financing for her expenditure
purposes. The corresponding impact of these elements for achieving accelerated economic growth and poverty
reduction is important. Therefore, it is prudent to investigate how financing of public expenditure foreign aid
(grants) and domestic revenue generation could be done optimally to help reduce poverty in Ghana.

3.0 Literature Review
The relationship between foreign aid and economic growth has been studied for over three decades. These
studies have concentrated on micro and macro levels with varying results. While some of the studies have found
direct positive relationship, others propose non-linear relationship. In the same vain, the numerous studies have
been undertaken with different methods. Also, the studies have generally looked at the effectiveness of aid
through its effect on fiscal policy on one hand and the impact of aid on economic growth on the other. These
differences warrant country specific study, relatively on macro-level to ascertain the impact of foreign aid on
economic growth.
The results of the previous studies have provided varied positive impact. Burnside and Dollar (2000) by
0
10
20
30
Figure 2: Trend of Aid and Domestic Revenue As Percentage of GDP
Aid As Percentage of GDP Domestic Revenue As Percentage of GDP
-15
-10
-5
0
5
10
15
Figure 1: Real GDP Growth Rates of Ghana
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
27
incorporating aid-policy interaction in their model found non-linear relationship. For them, foreign aid
contributes to economic growth so long as there is existence of good policy environment while Doucouliagos
and Paldam (2008) shows aid has no optimism for growth. In a similar way, Hadjimichael et al. (1995), Durbarry
et al. (1998) and Lensink and White (2001) found that there is a threshold effect of aid on growth. In other words,
aid exhibits a decreasing returns impact on output growth. Mosley (1987) found positive effect of aid at the
micro level but not at the macro, a situation termed as micro-macro paradox. The effectiveness of aid on growth
is premised on three main channels: aid increases domestic investment, aid increases aggregate savings in the
recipient countries and aid improves growth with some conditions (such as capital accumulation led growth,
good policy environment, etc). Earlier studies such as that of Boone (1996) showed that aid has no impact on
either investment or growth when viewed from neo-classical growth paradigm. That study was specified as linear.
On the other hand, other studies have results which contradict that of Boone. Studies such as Hadjimichael et al.
(1995), Burnside and Dollar (2000) and Lensink and White (2001) modeled their growth model as non-linear as
well as specifying foreign aid as a fraction of GDP. In spite of these similarities among non-linear specifications,
Boone included aid as an interactive term while others incorporated aid and its square as explanatory variables.
In addition, there have been criticisms about the methodology used. Bivariate Granger causality tests such as in
Dhakel et al. (1996) failed to establish any causal link between the two macroeconomic variables. Contrary to
the above studies, those who found positive link of aid to growth include Levy (1988), Murty et al. (1994) and
Grounder (2001). There have been issues raised about the endogeneity of aid flows as emanating from good
macroeconomic policy of recipient countries, income level of donor countries, etc. But evidence shows that there
are uncertainties surrounding aid flows including global financial and oil crisis, economic changes in donor
countries, etc. These among others have interrupted a consistent determination of aid flows. Based on these, aid
is treated as exogenous in this study.
3.1 Conceptual Framework and Model Specification
Recent aid literature centres on two competing models: a) the Good Policy Model, where the key feature is
policy interaction with aid (i.e. conditional on good policy), and b) the Medicine Model, where aid is squared
(Jensen and Paldam, 2006). This study is situated within the medicine model, basically because the policy
environment does not seems to be a major concern unlike the absorptive capacity of aid as evidenced by Foster
and Killick (2006), and Chowdhury and McKinley (2007). The medicine model reflects the fact that, there can
be a direct effect from aid to growth as well as acting through capital accumulation (Herzer and Morrissey, 2011),
and savings and investment (Hadjimichael et at, 1995; Hansen and Tarp, 2000, 2001). This non-linear theory
presents three scenarios:
The case of positive (increasing) effect of foreign aid on growth;
The case of maximum effect of foreign aid on output-optimal threshold; and
The case of decreasing effect of foreign aid on growth.
These cases present an inverted U-shape curve called Aid Laffer curve (Figure 3).
Output Positive Effect [
uP
Ad
> 0] No Effect [
uP
Ad
= 0]

Negative Effect [
uP
Ad
u]




Foreign Aid
Figure 3: Graph of Aid Laffer Curve
There is a general need for an initial foreign capital to supplement domestic savings and revenue. Thus, in theory,
below a certain threshold, acquisition of foreign capital is a spur to growth. This is referred to as the Stimulus View.
It argues that, with correct levels of aid resources, investment, employment and consumption, economic growth
would be stimulated. It is assumed that, this level of accumulation is helpful and serves as a catalyst to growth.
This positive relationship between aid and growth is based on the fact that, the new foreign capitals in the form of
grants afford the recipient country the ability to import critical capital goods and improve investment to enhance
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
28
growth. In the context of the theoretical underpinnings of the non-linear relationship, some reasons have been put
forward. For instance, the non-linear impact of aid on growth according to Lensink and White is due to the
existence of and inappropriate use of technology with institutional weaknesses. For Durbarry et al. capacity
constraints and elements of Dutch disease of aid are reasons while Hadjimichael et al. relate the non-linearity to
recipient countrys weak absorptive capacity problems.
Theoretically, domestic revenue raised through mainly taxes would have a non-linear relationship with economic
growth. The other component of domestic resources mobilisation is non-tax sources. The relative magnitude of
these in revenue, the rate of increase of tax rates, the mix of direct and indirect taxes, the size of the tax base and
how the tax base is expanding would determine whether non-linear effect of domestic revenue is possible. For this
study, the non-linear consideration is ignored based on a number of reasons. Firstly, for Ghana, tax rates are
marginally reduced or maintained through annual government budgets statements. Secondly, tax structure and
base are continually diversified and thirdly annual targets of revenue generation are exceeded. This to a large
extent indicates that, non-linear tax rates can be ignored.
In addition to aid and domestic revenue variables, a number of other variables are included as control variables.
The growth model estimated is standard according to literature. The control variables included traditional factors
of production: labour force (LF), gross capital formation (GCF) and human capital development (H) expected to be
positively related to economic growth; macroeconomic stability captured by inflation (IF) and the external sector
represented by openness (trade intensity-OP) indicating trade policy (Javid and Qayyum, 2011). The
macroeconomic variables of concern are real GDP growth rate (), foreign aid (A) and domestic revenue (DR).
Using secondary data on Ghana specific case and empirical literature, an econometric model is built to help
appreciate foreign aid and domestic revenue interaction and effect on output growth. The empirical model is
derived from the theoretical and empirical formulations. The objectives of the study are considered by using least
square estimation and Granger causality tests. Thus, the growth model becomes:
( ) ( ) ( ) ( ) ( ) ( ) ( )
2
0 1 2 3 4 5 6 7 8 9 t t t
t t t t t t t
y LF GCF H A A OP IF DR Dum

= + + + + + + + + + +
(1)
where
i
are the coefficients and is the error term.
Note:
y

: Real GDP growth rate (source-Ghana Statistical Service, 2012)

LF: Labour force growth rate (source-Ghana Statistical Service, 2012); GCF: Gross capital formation as
percentage of GDP (World Development Indicators, 2012); A: Foreign Aid as percentage of GDP (source-Bank
of Ghana, 2010 and World Development Indicators, 2012); DR: Domestic Revenue as percentage of GDP; H:
Public education expenditure as a percentage of GDP (source-ISSER, Ghana Statistical Service, 2012); OP:
Level of openness (source- World Development Indicators, 2012); IF: Inflation rate (source-ISSER, various
issues and Bank of Ghana, 2012); Dum: Form of governance and institutions. The dummy takes the value of 1
for all years in which there were democratically elected government and the value of zero for any other year.

4.0 Data Analysis and Results
The secondary data for the study is sourced from Bank of Ghana, Ghana Statistical Service and World
Development Indicators, covering the period 1970 to 2011. Generally, the time series variables are weighted by
GDP. The essence of regression analysis is to estimate short and/or long term economic relationships. Such
analyses are to test theoretical economic postulations. But recent developments in econometrics reveal the need
to ascertain the stationarity of time series properties. The general view is that, all variables should be stationary
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
29
at the same order of integration before estimation is undertaken (Adam, 1992 and Egwaikhide, 1997).
Unit root tests were performed. This was done using two recent improvements in the unit roots tests: Augmented
Dickey-Fuller and the Elliott-Rothenberg-Stock DF-GLS tests. The results of these tests are presented in Table 1.
Two main approaches are used to ascertain cointegration in the study. These are the two-step Engle-Granger
(1987) approach and Johansen method. This allows for a distinction to be made between the short and long runs.
If the growth rate, external aid, domestic revenue and other control variables are integrated of order 1, then in
order to ensure stationarity, the dependent variable should be expressed as a first difference, y
t
. When all
variables are I(1), and they are co-integrated, then non-stationarity is not an issue (Jenkins, 1998: pp. 42).
Table 1: Unit Root Test Results
DF-GLS Test Augmented Dickey-Fuller Test
Variables DF-GLS Statistic Comment Augmented DF Statistic Comment
y 3.36 I(0) -7.27 I(0)
LF -1.80 I(1) -2.58 I(1)
LF -2.24 I(0) -8.23 I(0)
GCF -0.48 I(1) -0.65 I(1)
GCF -7.19 I(0) -7.32 I(0)
H -2.38 I(0) -2.36 I(1)
H -5.49 I(0) -5.59 I(0)
A -1.58 I(1) -1.92 I(1)
A -8.81 I(0) -8.76 I(0)
DR -1.11 I(1) -1.15 I(1)
DR -2.61 I(0) -6.39 I(0)
IF -2.11 I(0) -2.58 I(1)
IF -11.07 I(0) -8.23 I(0)
OP -0.94 I(1) -0.94 I(1)
OP -6.06 I(0) -6.90 I(0)
The critical values for the DF-GLS
are -2.65 (for 1%), -1.94 (for 5%)
and -1.61 (for 10%)
The critical values for the Augmented DF test are
-3.60 (for 1%); -2.93 (for 5%) and -2.60 (for 10%)

It can be noted from Table 1 that, the first difference of the variables are all significant at 1% level implying that
they are stationary at first difference. This gives way for cointegration test to be performed, ensuring that there is
a long run relationship between the dependent variable on one hand and the independent variables on the other.
The cointegration tests were performed using two main approaches: the Engle-Granger and the Johansen
cointegration tests. The Engle-Granger cointegration test was based on the error term extracted after estimating
the long run static model in equation 1. If the unit root test on the error terms from the static model is stationary,
there is cointegration, which indicates that the model can be estimated in first difference.
The Engle-Granger two step cointegration test was applied along side the Johansen cointegration test. From
Table 2, the unit root test conducted on the error term extracted from the long run static model shows stationarity,
even at conventional levels of 1% and 5%. This indicates that, there is long run relationship among the variables
of the economic growth model.

Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
30
Table 2: Engle Granger Cointegration Test Results
Lag Length: 5 (Automatic based on SIC, MAXLAG=9)
Null Hypothesis: ECT has a unit root

Exogenous Assumptions
Constant
Constant, Linear
Trend
None
t-Statistic Prob. t-Statistic Prob. t-statistic Prob.
Augmented Dickey-Fuller test statistic -3.9645 0.0044 -3.9314 0.0214 -4.0286 0.0002
Test critical values: 1% level -3.6394 -4.2528 -2.6347
5% level -2.9511 -3.5484 -1.9510
10% level -2.6143 -3.2070 -1.6109
*MacKinnon (1996) one-sided p-values
The Johansen cointegration test summary was carried out and the results are presented in the Table 3 below. With
various assumptions experimented about the variables, there is at least one cointegration relationship among the
variables. It is noted that, with the trace statistics, there is cointegration irrespective of the test type. On the other
hand, the Maximum-Eigen Values show cointegration for the variables.
Table 3: Johansen Cointegration Test Results
Selected (0.05 level*) Number of Cointegrating Relations by Model
Data Trend: None None Linear Linear Quadratic
Test Type No Intercept Intercept Intercept Intercept Intercept
No Trend No Trend No Trend Trend Trend
Trace 2 3 3 3 4
Max-Eig 2 2 2 1 1
*Critical values based on MacKinnon-Haug-Michelis (1999)
Having established cointegration
1
, the study estimated an encompassing parsimonious model. Given this
situation, then a dynamic model of growth can be built by creating an error correction term (ECT) which
reflects the convergence of GDP growth on its long run equilibrium path. Along with the contemporaneous
values of the explanatory variables, their lagged values were also added as separate variables. Equally, the lagged
values of the dependent variable were included. This is based on the fact that, some of the variables have impact
both in the current and future periods. Their effects transcend beyond one period. An arbitrary lag length of three
was initially used. The insignificant variables were eliminated by using the standard Akaike and Scharwz
Information Criteria. This is a General-to-Specific approach. The coefficient of ECT measures the speed with
which GDP growth converges on its long run equilibrium. It is expected to be a) negative; b) less than unity and
c) statistically significant. The other variables in the equation will be those which capture or determine GDP
growth in the short run. Changes in GDP growth will thus be a function not only of current changes in the
independent variables, but also of the cumulative effect of lagged changes in all the variables and the adjustment
of total GDP towards its long-run equilibrium path (indicated by the error correction term). The general form of
the error correction model of growth can be given as:
1 1 1 1 1
2
0 1 2 3 4 5
0 0 0 0 0
n n n n n
t i t i t i t i t i t
i i i i i
y LF GCF H A A

= = = = =
= + + + + +



1
Two series are cointegrated if they never drift apart from each other, that is, they maintain
equilibrium. That is a linear function of the two is I(0), stationary
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
31
1 1 1
6 7 8 9 10 1
0 0 0
n n n
t i t i t i t t
i i i
OP IF DR Dum ECT


= = =
+ + + + + +


The error correction model captures the short run dynamic as well as long run properties of the economic growth
model. This is because, it includes variables both in levels (i.e., error correction term, dummy) and differences
(of all other explanatory variables). Therefore, the model can be estimated with Least Square method since all
the variables included in the model are stationary. By applying the general-to-specific methodology, the
parsimonious error correction model results are presented in Table 4 below.
Table 4: Results of the Short Run Models of Economic Growth
Variables Coefficients of Short Run Models
General Model Model Without
Foreign Aid
Model Without
Domestic
Revenue
Model Without Both
Domestic Revenue and
Foreign Aid
(i) (ii) (iii) (iv)
Constant 0.10 (1.14) 0.63 (1.71) 0.10 (1.19) 0.510 (2.56)
Foreign Aid 0.57 (2.40)b

0.59 (2.23)b
Foreign Aid
Square
-0.007 (-2.16) -0.007 (-1.40)
Domestic
Revenue
0.65 (3.05)b 0.39 (2.41)b
Inflation 0.03 (2.23)b 0.036 (199)a 0.03 (1.92)a 0.03 (1.86)a
Gross Capital
Formation
0.57 (3.70)c 0.45 (2.27)c 0.40 (1.96)c 0.50 (2.52)c
Human Capital
Development
0.58 (4.73)c 0.42 (2.97)c 0.70 (1.85)a 0.13 (2.66)c
Labour Force 0.35 (1.39)c 0.23 (1.18)b 0.20 (1.54)a 0.29 (1.37)a
Openness 0.14 (2.73)c 0.09 (1.37)c 0.07 (1.88) 0.11 (1.71)c
Dummy 0.45 (1.48) 0.33 (1.10) 0.56 (1.4) 0.21 (1.99)
Error Correction
Term
-0.23 (-6.71)a -0.54 (-3.09) -0.59 (1.5)a -0.59 (-3.39)

Adjusted R-Sq. 0.74 0.64 0.63 0.61
Prob (F-Statistic) 0.00 0.00 0.00 0.00
Note: t-statistics in parenthesis; a, b, and c are lags 1, 2 and 3 respectively
The general model demonstrates the impact of foreign grants and domestic revenue as postulated by theory.
Foreign grants show a nonlinear effect on economic growth. The level variable is positive while its square gives
a negative effect (equation i). This shows that, GDP growth initially increases as aid increases but falls after
some point. The combined effect indicates that, increasing acquisition of foreign aid has a threshold in its
contribution to output growth. Thus, given the results, aid threshold is estimated to be 40.7% of GDP. This is
attested to by other studies such as Chowdhury and McKinley (2007). Domestic revenue also has a positive
relationship with output growth. When the marginal diminishing effect of aid is dropped, the level variables
indicate that, domestic revenue contribution to output growth far outweighs the contribution of foreign grants.
When domestic revenue alone is excluded from the model, the impact of aid is marginally increased (see
equation iii) compared with the general model. This is because government investment largely depends on
domestic revenue. In spite of the fact that government of Ghana receives about 25% of budget support through
grants, domestic revenue while directly contributing to GDP also indirectly enhances the effect of money
received from abroad. Without the support of domestic revenue, grants may not be enough to undertake major
development activities. Also, without the inclusion of domestic revenue, the diminishing effect of aid becomes
non-existent over the study period. In addition, there is marginal change in the coefficient of aid when domestic
revenue is excluded (and when the level variable of aid alone is used), indicating that without domestic revenue,
there will be no problem with large amount of aid coming into the country. Thus, controlling for domestic
revenue, the diminishing effect of aid does not apply in Ghana over the study period (as it is insignificant).
Without the inclusion of foreign aid, domestic revenue still maintained a positive effect on growth. The
coefficient of domestic revenue reduced from 0.65 to 0.39 in the case where there is aid and where there is no aid
respectively. This implies that, the two financing components of government funding are complementary to each
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
32
other. This together with earlier results shows that, domestic revenue is a necessary source of funding for public
investment and development. At the extreme, domestic revenue could be substituted for aid but not vice versa
over an extended period.
The behaviour of the control variables are discussed via the inclusion and exclusion of the two main sources of
funding-grants and domestic revenue (equations ii and iii). The effect of inflation on growth was positive in all
the different versions of the model. Ironically, there was no much difference in the magnitude of the coefficients
but was relatively significant in the general model. These results show that, the price levels have not reached a
point that would hamper economic growth. Gross capital formation produced marked difference in the results in
the two cases. In the presence of the two funding sources, the coefficient was 0.57 but reduced to 0.13 when the
two variables of concern were excluded. This is indicative of the fact that, it serves as one of the major sources
of capital accumulation when there is no domestic revenue and foreign grants. Human capital development also
maintained positive relationship in the models as expected. Unlike inflation, its impact reduced from the case
where aid and revenue were included to the case where they were excluded (i.e. from 0.58 to 0.13). In this
situation, it can be deduced that, domestic revenue and aid help to improve human capital development in the
form of accessible quality education and providing health care services on affordable basis. Notably, Ghana
spends quite an amount of resources on human capital development subsector.
The expected positive impact of labour force on output growth was revealed. The impact reduced from the case
where the two variables were included to where they were not included. This may be due to the condition that,
labour becomes useful when capital accumulation constraints are non existent. In addition, there were marginal
changes in the contribution of trade intensity from 0.14 to 0.11. This may be due to the fact that, there is limited
link between trade openness on one hand and the receipt of foreign grants and domestic revenue generation on
the other. The objective of the study was also analysed from the perspective of Granger causality tests. The
unconditional pair-wise Granger causality tests results among foreign aid, domestic revenue and economic
growth are given in Table 5 below.
Table 5: Pairwise Granger Causality Tests
Null Hypothesis, Ho F-Statistics Probability Decision
A does not Granger Cause GDP 3.4077 0.017 Reject Ho
GDP does not Granger Cause A 1.009 0.432 Accept Ho
DR does not Granger Cause GDP 4.543 0.004 Reject Ho
GDP does not Granger Cause DR 2.741 0.041 Reject Ho
DR does not Granger Cause A 0.546 0.739 Accept Ho
A does not Granger Cause DR 2.006 0.112 Accept Ho
It is observed that, foreign aid Granger causes GDP growth. This shows that at 5% significant level, aid helps
GDP to grow. On the other hand, the hypothesis that GDP growth does not Granger cause A is accepted. These
show and confirm that, aid is given to recipient countries to help enhance GDP growth. Aid is given to improve
capital accumulation, complement human capacity, etc. In addition, there is a positive correlation between GDP
growth and foreign aid, of a value of 0.36. The causal link between domestic revenue and GDP is intuitive.
It is noted that, there is bi-causality between the two variables. At 1% significant level, domestic revenue
Granger causes GDP, while at 5%, GDP growth Granger causes domestic revenue. Thus, there is a stronger
causal link from domestic revenue to GDP growth. As more revenue is raised, government obtains more
resources which can enhance GDP growth. It also becomes advantageous to raise more revenue when GDP
growth is increasing.
Theoretically, domestic revenue generation is not supposed to influence the flow of foreign aid (except under
conditions of matching funds). This is confirmed in the Granger causality test. It is noted that, both hypothesis
are accepted: domestic revenue does not Granger causes foreign aid flow and vice versa. In Ghanas case,
foreign aid has mostly come to complement government budget. According to MOFEP (2009), annual foreign
aid budget support to Ghana in the 2000s has been around 25%. Even in the case of matching funds, domestic
revenue has to be raised. Domestic revenue seems more significant in determining the growth of GDP in Ghana
relative to foreign grants. In the context of magnitude, in the budget, the proportion of external funding in the
annual budget average 25%, indicating the remaining 75% of the funding is generated domestically. In this case,
assuming that each unit of foreign and domestic funding yields the same impact on output, then domestic
revenue would generate much larger contribution to GDP growth over time.

5.0 Conclusion
Various funding sources are available to any government in a developing country. In spite of their availability,
each source has its unique way of contributing to economic growth. External debt accumulation has over the
Journal of Economics and Sustainable Development www.iiste.org
ISSN 2222-1700 (Paper) ISSN 2222-2855 (Online)
Vol.4, No.8, 2013
33
years been problematic for developing countries. On the other hand, domestic revenue and foreign grants seem
to be funding sources that do not have immediate negative effect on output growth. This formed the basis for this
study. The results show that, domestic revenue and grants complement each other for development financing.
Moreover, grants cannot be a substitute for domestic revenue generation as domestic revenue is the most
important between the two factors. It was also noted that, when there are various forms of capital accumulation,
the threshold level of aid is pushed higher but aid becomes very effective in the absence of capital accumulation.
In addition, the threshold effect of aid becomes non-existent when there is no capital accumulation. Domestic
revenue on the other hand is effective irrespective of whether there is capital accumulation or not.
Also, it is noted that, domestic revenue and foreign aid are not well related over the study period. In other words,
the flow of foreign aid does not depend on whether there is enough domestic revenue or not. Contrary to this,
there is a causal link from both domestic revenue and foreign aid to economic growth. Moreover, both domestic
revenue and foreign aid have positive impact on economic growth. Given institutions and absorptive capacity,
there seem to be a threshold effect of aid on economic growth in Ghana. It is concluded that, both domestic
revenue and foreign aid are important for enhancing economic growth, but more attention should be given to the
generation of domestic revenue. However, these results should be taken with caution as issues of corruption and
mismanagement of fund are critical in the use of funds.

References
Adam, C. S. (1992). Recent Development in Econometric Methods An Application to the Demand for Money
in Kenya, AERC Research Paper 15, Zakuna Printers Ltd, Nairobi.
Boone, P. (1994). The impact of foreign aid on savings and growth. London School of Economics CEP
Working Paper (677).
Burnside, C. and D. Dollar (2000). Aid, Policies and Growth. American Economic Review. 90: 847- 868.
Chowdhury, A. and McKinley T. (2007). Gearing Macroeconomic Policies to Manage Large Inflows of ODA
The Implications for HIV/AIDS Programmes, UNU-WIDER, Research Paper No. 2007/43.
Doucouliagos, H and Paldam, M (2008). Aid effectiveness on growth: A meta study. European Journal of
Political Economy, 24: 1-24.
Durbarry, R., N. Gemmell, and D. Greenaway (1998). New evidence on the impact of foreign aid on economic
growth. CREDIT Research Paper 98/8, University of Nottingham.
Egwaikhide, F. O. (1997). Effects of Budget Deficits on the Current Account Balance in Nigerian: A Simulation
Exercise, African Economic Research Consortium Research paper 70, Regal Press Kenya Ltd, Nairobi.
Foster, M. and Killick, T. (2006). What would doubling aid do to macroeconomic management in Africa, ODI,
Working Paper 264, April.
Hadjimichael, M. T., D. Ghura, M. Mu hleisen, R. Nord, and E. M. Ucer (1995). Sub-Saharan Africa: growth,
savings, and investment. 1986-93. Occasional Paper 118, International Monetary Fund.
Hansen, H. and Tarp, F. (2000). Aid and growth regressions. Journal of Development Economics, 64: 547-570.
(2001). Aid and growth regressions. Journal of Development Economics, 64 (2),
Herzer, D and Morrissey, O (2011). Long-run aid effectiveness. Paper 144 presented at the CSAE Conference
2011: Economic Development in Africa, 20-22 March, Oxford.
Javid, M and Qayyum, A. (2011). Foreign aid-growth nexus in Pakistan: role of macroeconomic policies. MPRA
Paper 29498, Munich, MPRA.
Jenkins, C. (1998). Determinants of private investment in Zimbabwe. Journal of African Economies, Vol. 7, Issue
1
Jensen, P. S. and Paldam, M. (2006). Can the Two new aid-growth model be replicated? Public Choice, Vol.
127, No. 1-2, 147-175.
Lensink, R. and H. White (2001). Are there negative returns to aid? Journal of Development Studies,
37
Levy, V. (1988). Aid and growth in Sub-Saharan Africa: the recent experience. European Economic Review. 32:
Ministry of Finance and Economic Planning (2009). Annual External Financing Performance Report, MOFEP,
Accra, Ghana.
Mosley, P. (1987). Overseas Aid: Its Defense and Reform. Brighton: Wheatshead Books.
Murty, V. N. R., V. Ukpolo, and J. M. Mbaku (1994). Foreign aid and economic growth in Cameroon: evidence
from cointegration tests. Applied Economics Letters. 1: 161-163.
UNDP (2007). Human Development Report, Washington DC.